Question

下面是我的.csv文件的简要示例：

"X","Y","Z","N"
"X","Y","Z","N"
"X","Y","","N"
"X","Y","Z","N"
"X","Y","Z","N"
"X","Y","","N"
"X","Y","Z","N"
"X","Y","Z","N"
"X","Y","","N"

在第三列（“ Z”列）中，我缺少一些单元格（第3,6和9行）。最好使用awk或sed，我想专门针对第3列，如果任何单元格为空白，我想删除整行。我的最终结果将是：

"X","Y","Z","N"
"X","Y","Z","N"
"X","Y","Z","N"
"X","Y","Z","N"
"X","Y","Z","N"
"X","Y","Z","N"

对于我的实际项目，以下是我的一个文件的精确副本- https://github.com/drphillgood/riotapidata/blob/master/csv/game3.csv。您将在第28列（participants__participantId）中看到只有某些单元格具有数据（与最后一列的数据相同，参与者__playerName）。如果此列中的单元格之一为空白，我想使用.sh脚本删除整行。最终文件会这样-https://github.com/drphillgood/riotapidata/blob/master/csv/game3_v2.csv

Answer 1

一个更简单的AWK命令：

awk -F , '$3 != "\"\"" {print}' inputfile > outputfile

将字段分隔符设置为逗号，并打印出第三行仅包含""的每一行。

此功能不足以处理字段中包含逗号的CSV文件。它希望空字段由空引号组成。

Answer 2

这是一个awk脚本，可以完成操作。

awk -F '","' '!$3{next}1' input.csv

输出：

"X","Y","Z","N"
"X","Y","Z","N"
"X","Y","Z","N"
"X","Y","Z","N"
"X","Y","Z","N"
"X","Y","Z","N"

说明：

BEGIN {FS = "\",\""}  # input line field separator ","
!$3{next}             # if empty string in 3rd input field, skip
1                     # print current line

更新：

以下是注释，并在链接中提供了测试CSV文件。

测试$ 28字段

awk -F '","' '!$28{next}1' input.txt | awk -F '","' '{print $28}'的输出

participants__participantId
1
2
3
4
5
6
7
8
9
10

Answer 3

可以用sed命令完成： sed -r -n '/^([^,]*,){27}""/! p' yourfile

对于完整文件，请使用27；对于最小示例，请使用2：在需要检查的列之前中指定字段数。

正则表达式打印（p和-n选项）的行（/.../!表示否定否）与条件不匹配：

^匹配
([^,]*,)[27}由逗号分隔的27个字段可能会退化为27个逗号，
""仅在下一个字段中使用双引号

Answer 4

awk -F, '$3 ~ /"Z"/{print $0}' file

"X","Y","Z","N"
"X","Y","Z","N"
"X","Y","Z","N"
"X","Y","Z","N"
"X","Y","Z","N"
"X","Y","Z","N"

使用bash（awk / sed）在第28列中没有数据时删除行

4 个答案:

更新：